The Host Defense Peptide Beta-Defensin 1 Confers Protection against Bordetella pertussis in Newborn...

10.1128/IAI.74.4.2338-2352.2006.

2006, 74(4):2338. DOI:Infect. Immun. Hugh G. G. Townsend, Lorne A. Babiuk and Volker GerdtsShokrollah Elahi, Rachelle M. Buchanan, Sam Attah-Poku,

in Newborn PigletspertussisBordetellaConfers Protection against

The Host Defense Peptide Beta-Defensin 1

http://iai.asm.org/content/74/4/2338Updated information and services can be found at:

These include:

REFERENCEShttp://iai.asm.org/content/74/4/2338#ref-list-1at:

This article cites 57 articles, 31 of which can be accessed free

CONTENT ALERTS more»articles cite this article),

Receive: RSS Feeds, eTOCs, free email alerts (when new

http://journals.asm.org/site/misc/reprints.xhtmlInformation about commercial reprint orders: http://journals.asm.org/site/subscriptions/To subscribe to to another ASM Journal go to:

on May 13, 2014 by guest

http://iai.asm.org/

Dow

nloaded from


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/content/74/4/2338

http://iai.asm.org/content/74/4/2338#ref-list-1

http://iai.asm.org/cgi/alerts

http://journals.asm.org/site/misc/reprints.xhtml

http://journals.asm.org/site/subscriptions/

http://iai.asm.org/

http://iai.asm.org/

INFECTION AND IMMUNITY, Apr. 2006, p. 2338–2352 Vol. 74, No. 40019-9567/06/$08.00�0 doi:10.1128/IAI.74.4.2338–2352.2006Copyright © 2006, American Society for Microbiology. All Rights Reserved.

The Host Defense Peptide Beta-Defensin 1 Confers Protectionagainst Bordetella pertussis in Newborn Piglets

Shokrollah Elahi, Rachelle M. Buchanan, Sam Attah-Poku, Hugh G. G. Townsend,Lorne A. Babiuk, and Volker Gerdts*

Vaccine and Infectious Disease Organization, University of Saskatchewan, Saskatoon, SK S7N 5E3, Canada

Received 2 November 2005/Returned for modification 21 December 2005/Accepted 13 January 2006

Innate immunity plays an important role in protection against respiratory infections in humans andanimals. Host defense peptides such as beta-defensins represent major components of innate immunity. Werecently developed a novel porcine model of pertussis, an important respiratory disease of young children andinfants worldwide. Here, we investigated the role of porcine beta-defensin 1 (pBD-1), a porcine defensinhomologue of human beta-defensin 2, in conferring protection against respiratory infection with Bordetellapertussis. In this model, newborn piglets were fully susceptible to infection and developed severe bronchopneu-monia. In contrast, piglets older than 4 weeks of age were protected against infection with B. pertussis.Protection was associated with the expression of pBD-1 in the upper respiratory tract. In fact, pBD-1 expres-sion was developmentally regulated, and the absence of pBD-1 was thought to contribute to the increasedsusceptibility of newborn piglets to infection with B. pertussis. Bronchoalveolar lavage specimens collected fromolder animals as well as chemically synthesized pBD-1 displayed strong antimicrobial activity against B.pertussis in vitro. Furthermore, in vivo treatment of newborn piglets with only 500 �g pBD-1 at the time ofchallenge conferred protection against infection with B. pertussis. Interestingly, pBD-1 displayed no bacteri-cidal activity in vitro against Bordetella bronchiseptica, a closely related natural pathogen of pigs. Our resultsdemonstrate that host defense peptides play an important role in protection against pertussis and are essentialin modulating innate immune responses against respiratory infections.

Pertussis, an acute respiratory tract infection caused by Borde-tella pertussis, remains an epidemic disease responsible for signif-icant infant and child morbidity and mortality. The disease is stillperceived as a serious global health problem, with more than 45million cases and more than 400,000 deaths every year (13). Im-munization with either whole-cell or acellular vaccines has signif-icantly reduced the disease load in developed and developingcountries. However, the mechanisms of disease protection, inparticular, the role of innate immunity in protection, are poorlyunderstood. We recently developed a novel model of pertussis innewborn piglets and demonstrated that in contrast to newbornpiglets, older piglets (�4 weeks of age) were fully resistant toinfection with B. pertussis (15). Their response was both rapid andindependent of previous exposure, indicating that innate immu-nity was crucial for disease protection.

Epithelia of skin and mucosa both serve as a physical barrieragainst pathogens and produce a vast array of endogenoussubstances including antimicrobial agents that can inhibit orneutralize invading pathogens (56). Many of these elementsare the components of the innate immune response. The ef-fector branch of innate immunity consists of two major mech-anisms: the recruitment and activation of cellular componentsincluding macrophages, neutrophils, natural killer cells, anddendritic cells (DCs) and the release of a broad spectrum ofextracellular mediators such as cytokines, chemokines, com-plement, and antimicrobial peptides (AMPs) (4, 56).

AMPs, also called host defense peptides (HDPs), include awide range of proteins that can be classified into defensins andcathelicidins. HDPs have both direct, broad-spectrum antimi-crobial activity and the ability to modulate immune responsesagainst bacteria, fungi, parasites, and even viruses in a widevariety of species (24, 29, 35). Direct killing of bacteria byHDPs is mediated mainly through permeabilization of bacte-rial membranes (31, 58–60). Immunostimulatory functions in-clude the ability to induce chemotaxis of immature DCs and Tcells, activation of antigen-presenting cells, and induction ofglucocorticoid production, macrophage phagocytosis, mast celldegranulation, complement activation, and interleukin-8(IL-8) production by epithelial cells (7, 36, 57). Thus, HDPsappear to represent an important link between innate andacquired immunity and are potent immune modulators andadjuvants for vaccines (7). Beta-defensins, cationic peptideswith broad-spectrum antimicrobial activity, are widely ex-pressed in the skin and the mucosal surfaces of airways, thedigestive tract, and the urogenital tract (16, 26, 27, 53, 54).They are either constitutively expressed or induced by bacterialproducts or proinflammatory cytokines. As part of the innateimmune system, defensins are thought to play a significant rolein protection against respiratory infections (14, 19, 44, 55, 57).

Pigs, like many other mammals, produce an impressive arrayof HDPs, such as cathelicidins PR-39 and protegrin 1 (PG-1),and defensins, which are synthesized predominantly by hostphagocytes and mucosal epithelial cells. Although other beta-defensin genes have been identified (GenBank accession num-ber AY460575) (42), porcine beta-defensin 1 (pBD-1) is theonly well-studied beta-defensin with significant homology tohuman beta-defensins in pigs. pBD-1 is expressed throughout

* Corresponding author. Mailing address: Vaccine and InfectiousDisease Organization, University of Saskatchewan, Saskatoon, SK S7N5E3, Canada. Phone: (306) 966-1513. Fax: (306) 966-7478. E-mail: [email protected].

2338


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

the epithelia of the tongue, the respiratory and gastrointestinaltracts, and a variety of other organs. Interestingly, expressionhas not been detected in peripheral blood mononuclear cells(PBMCs) (46, 61). pBD-1 has direct antimicrobial activityagainst a variety of porcine pathogens including Listeria mono-cytogenes, Candida albicans, Escherichia coli, Streptococcus suis,and Actinobacillus pleuropneumoniae (46). Furthermore, it alsohas immunomodulatory activity in porcine PBMCs and lymphnode cells (data not shown).

Here, we demonstrate that resistance against respiratoryinfection with B. pertussis in older piglets was associated withthe presence of pBD-1 in the upper respiratory tract. Further-more, treatment with pBD-1 conferred complete protection ofnewborn piglets from intrapulmonary infection with B. pertussis.Thus, using a relevant model for a human disease of greatimportance, we show here that beta-defensins are involved indisease protection against respiratory infection. Our results notonly further advance the understanding of the immunopatho-genesis of pertussis but also show that treatment with defensinsmay represent an alternative strategy for treating pertussis inhuman patients.

MATERIALS AND METHODS

Bacterial cultures. B. pertussis suspensions of strain Tohama I were stored at�70°C in Casamino Acids plus 10% glycerol. Organisms were initially grown onBordet-Gengou (BG) agar (Becton Dickinson and Company) containing 15%(vol/vol) defibrinated sheep blood and 40 �g/ml of cephalexin (Sigma-Aldrich) at37°C for 48 h. Heavy inocula of bacteria were resuspended in Stainer-Scholte(SS) medium and grown aerobically at 37°C for 48 h either as a liquid culture at250 rpm in a Thermo Forma shaker or on BG plate cultures. Bacteria wereharvested from BG plates by scraping off and resuspending bacteria in SS me-dium. After centrifugation at 2,500 � g for 10 min, pellets were resuspended inMg2�- and Ca2�-free phosphate-buffered saline (PBSA) (pH 7.2, 20 mM) andadjusted to a specified bacterial titer (3 � 109 to 4 � 109 CFU) by determiningthe optical density at 600 nm (Ultrospec 3000; Pharmacia Biotech, United King-dom). The bacterial suspension (adjusted to 50% from liquid culture and 50%from BG plates) was kept on ice until it was used for challenge infection.Corresponding viable counts were determined by plating serial dilutions onto BGagar plates and subsequent incubation at 37°C for 4 to 5 days. For in vitroinhibition assays, bacteria were grown on BG agar, scraped off, washed in PBSA,and resuspended in SS medium at 5 � 108 to 7 � 108 CFU/ml. Bordetellabronchiseptica rabbit isolate RB50 (12, 43) (kindly provided by Andrew Preston,University of Guelph, Canada) was stored at �70°C in Casamino Acids plus 10%glycerol. Organisms were grown on BG agar plates (supplemented with 15%defibrinated sheep blood and cephalexin) at 37°C for 24 h. Bacteria were resus-pended in SS medium at 5 � 108 to 7 � 108 CFU/ml of SS medium for in vitroinhibition assays.

Animals. Pregnant Landrace sows were purchased from the Saskatoon PrairieSwine Centre, University of Saskatchewan. Sows were induced to farrow byintramuscular injection of prostaglandin (Planate) (Schering, Quebec, Canada)at day 113 of gestation. Piglets were born at day 114 to 115 of gestation. Nursingpiglets were kept within the same room but in separate pens and monitoredvery closely. The piglets were challenged at 3 to 5 days of age. All experimentswere conducted in accordance with the ethical guidelines of the University ofSaskatchewan and the Canadian Council for Animal Care.

Respiratory infection and postmortem analysis. Newborn piglets were anes-thetized with isoflurane and intubated using a laryngoscope. Infection was initi-ated by delivering 1.5 ml of PBSA (20 mM) containing 5 � 109 CFU of B.pertussis strain Tohama I intrapulmonarily (craniodorsal of the bifurcation)through a Micro-Renathane tube (size 0.95; Braintree Scientific Inc.), which wasinserted through an endotracheal tube (3 mm; Jorgensen Laboratories Inc.,Loveland, CO). The bottom end of the Micro-Renathane tube was sealed, andminute holes were made for the equal distribution of bacteria. Piglets weremonitored twice daily for clinical symptoms including fever and respiratorysymptoms such as nasal discharge, nonparoxysmal cough, and breathing difficul-ties. Piglets were euthanized by intraperitoneal injection of 5 ml of sodiumbarbiturate (Euthanyl; Bimeda-MTC, Ontario, Canada) at different time points

over a 10-day period after challenge. The thoracic and abdominal cavities and thelungs were examined for any lesions, and abnormalities such as pleuritis or localcollections of blood and fluids in the thorax were noted.

Isolation of bacteria. The number of bacteria in the bronchoalveolar lavage(BAL) specimens and lung lesions was examined over a 10-day period. To allowoptimal recovery, because B. pertussis is a fastidious bacterium, SS medium wasused to lavage the lungs. The BAL fluid was obtained by filling the lungs with 15ml of SS medium and withdrawing as much fluid as possible. To quantify thepresence of B. pertussis in the BAL fluid, samples were centrifuged to removedebris and host cells; supernatant and dilutions thereof were plated onto BGplates and incubated at 37°C for up to a week. To determine the number ofbacteria within the tissues, lesions were excised, weighed, homogenized, andplated onto BG agar plates.

In vivo treatment with pBD-1. Prior to challenge infection with B. pertussis, 500�g of pBD-1 in 1.5 ml of PBSA (20 mM) was delivered through the Micro-Penthane tube into the lungs of anesthetized piglets. In the early experiments, weused the same Micro-Penthane tube to deliver the peptide and bacteria. How-ever, we cleared the tube of any remaining peptide by passing 30 ml of airthrough the tube before the bacteria were delivered. In further animal experi-ments, to avoid possible direct contact of bacteria and the peptide, the Micro-Penthane tube was changed and a new tube was inserted for delivering thebacteria. Either method resulted in protection, and there was no significantdifference in the outcome of challenge. Control groups received 1.5 ml of PBSA.Following treatment, animals from both groups were infected with 5 � 109 CFUbacteria in 1.5 ml of PBSA (20 mM) as described above.

Reverse transcriptase PCR. Total RNA was extracted from porcine tissues bythe acid guanidium thiocyanate/phenol-chloroform method using a total RNAisolation reagent (TRIzol reagent; GIBCO BRL). RNA was reverse transcribedto cDNA using the oligo(dT) primer (Sigma-Aldrich) according to the manu-facturer’s instructions. Reverse transcriptase products were amplified using aPCR kit (RED Taq, Ready Mix PCR Mix; Sigma-Aldrich) and a thermal cycler(Gene Amp PCR system 9700; Applied Biosystems, Singapore). The primersused to detect a 287-bp cDNA sequence of pBD-1 in porcine tissues were senseprimer 5�-TCCCATGAGACTCCACCGCCTCCT-3� and antisense primer 5�-TTCGAGCAGCTTCTGAGCCATATCTGTAC-3�; primers used to detect a100-bp cDNA sequence of PG-1 were sense primer 5�-TGGATCAGATCAAGGACCTA-3� and antisense primer 5�-ACACAGACGCAGAACCTAC-3�; prim-ers used to detect a 262-bp cDNA sequence of PR-39 were sense primer 5�-ACCCATCCATTCACTCAC-3� and antisense primer 5�-AGCCACAACAATAAGATCC-3�; primers used to detect a 452-bp cDNA sequence of GAPDH(glyceraldehyde-3-phosphate dehydrogenase) were sense primer 5�-ACCACAGTCCATGCCATCAC-3� and antisense primer 5�-TCCACCACCCTGTTGCTG-3�. The PCR conditions were a 1-min denaturation step at 95°C followed by 24cycles of denaturation at 94°C for 15 s, annealing at 60°C for 30 s, and extensionat 72°C for 30 s, followed by a final extension step at 72°C for 7 min. The PCRproducts were visualized by electrophoresis on 1.5% agarose gels containing 0.5�g/ml ethidium bromide.

In vitro growth inhibition assays. Noninfected piglets, either 4 to 5 weeks oldor newborn (6-h-old colostrum-deprived or colostrum-fed piglets), were eutha-nized, and BAL specimens were collected in 20 mM PBSA, which was used tofacilitate optimal conditions for the AMPs. The BAL fluid was obtained by fillingthe lungs with 10 ml of PBSA and withdrawing as much fluid as possible (thisprocedure was performed once). Alveolar macrophages and other cells wereremoved by centrifugation at 500 � g for 10 min. BAL specimens (290 �l) werecocultured in microtiter plates with 10 �l of a bacterial suspension containing5 � 106 to 7 � 106 CFU of B. pertussis at 37°C. Supernatants were plated ontoBG agar plates at different time points to evaluate the number of viable bacteria.The sensitivity of B. pertussis to human lysozyme (Sigma), human lactoferrin(Sigma), synthetically derived pBD-1, human beta-defensin 2 (hBD-2), and PG-1was assayed by coculturing appropriate concentrations of lysozyme, lactoferrin,pBD-1, hBD-2, and PG-1 in 20 mM PBSA (280 �l) with 5 � 106 to 7 � 106 CFU(10 �l) bacteria. Plates were incubated at 37°C for 1, 2, and 4 h. The sameprocedures were used to investigate the inhibitory effects of BAL fluid, pBD-1,hBD-2, and PG-1 against B. bronchiseptica in vitro.

Synthesis of the pBD-1, hBD-2, and PG-1. pBD-1, hBD-2, and PG-1 werechemically synthesized on a Pioneer solid-phase peptide synthesizer (PerSep-tive Biosystems, Foster City, CA) using Fmoc (9-fluorenylmethoxy carbonyl)chemistry. The peptide chains were synthesized from the carboxyl terminus tothe amino terminus onto [5-(4-Fmoc-aminomethyl-3,5-dimethyloxyphenoxy)valeric acid]–polyethylene glycol–polystyrene (PAL-PEG-PS) resin. BothFmoc-protecting groups at the amino terminus were deprotected with piper-idine. The peptides were cleaved from the resin with concurrent deprotectionof the side chain-protecting groups by treating the resin-bound peptide with

VOL. 74, 2006 BETA-DEFENSIN 1 PROTECTS AGAINST PERTUSSIS IN PIGLETS 2339


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

trifluoroacetic acid (TFA) (9.3 parts) in the presence of scavengers (anisole–ethyl-methyl sulfide–1,2-ethanedithiol [3:3:1]), for 7 h. The crude peptideswere filtered from the resin, and the TFA was evaporated. Diethyl ether wasadded to the residues to precipitate the crude peptide. The peptides wereisolated and purified by high-performance liquid chromatography (HPLC) onVydac protein C4 columns (1.0 by 25 cm) eluting with a linear gradient of35% buffer A (H2O–0.1% TFA)–90% buffer B (acetonitrile–H2O [90/10]–0.01% TFA) for 40 min at a flow rate of 3 ml/minute. The purity andmolecular weight of the respective peptides were confirmed by matrix-as-sisted laser desorption ionization (MALDI)–time of flight mass spectrometryon a PE Biosystems Voyager system 4068 (National Research Council, PlantBiotechnology Institute, Saskatoon, Canada) and by amino acid analysis.

The cysteine residues were oxidized in 0.1 M ammonium acetate buffer (pH7.7) in the presence of reduced and oxidized glutathione at a ratio of 1/20/2(peptide/reduced glutathione/oxidized glutathione) as described previously byHiratsuka et al. (30). The oxidized peptides were purified by high-performanceliquid chromatography on a Vydac protein C4 column (1.0 by 25 cm) eluting witha linear gradient of 35% buffer A (H2O–0.1% TFA)–90% buffer B (acetonitrile–H2O [90/10]–0.01% TFA) for 40 min at a flow rate of 3 ml/minute. Since we wereunable to perform two-dimensional nuclear magnetic resonance on the synthe-sized peptides to determine the correct pairing of the intermolecular disulfidebonds, we compared the bactericidal activity of synthetic pBD-1, hBD-2, andPG-1 against E. coli to the previously reported activity of a recombinant baculo-virus-expressed peptide (30, 46, 48). Furthermore, BAL analysis provided strongevidence for synthetic pBD-1 and PG-1 being correctly folded, as both migratedsimilarly on reverse-phase HPLC (RP-HPLC) to their corresponding naturalpeptides. Also, MALDI data for pBD-1 confirmed the loss of 6 mass units,indicating that three disulfide bridges were formed.

Production of anti-pBD-1 antibody. Rabbit anti-pBD-1 serum was producedby immunization with synthetic pBD-1. The peptide was conjugated to eitherovalbumin or keyhole limpet hemocyanin by using a single-step cross-linkingtechnique with glutaraldehyde (28). New Zealand White rabbits were immunizedwith conjugated pBD-1 (300 �g) in complete Freund’s adjuvant in a total volumeof 1.25 ml subcutaneously at multiple sites. Booster immunizations were per-formed at 2 and 4 weeks after priming, blood was drawn 10 days after the lastbooster, and serum was analyzed by Western blot and stored at �20°C.

Transmission electron microscopy. Transmission electron microscopy was em-ployed to evaluate the effect of pBD-1 on the bacterial membranes of B. pertussisand B. bronchiseptica. pBD-1 was added at a final concentration of 20 �g/ml to5 � 109 to 7 � 109 CFU bacteria and incubated for 20 to 60 min at 37°C. Bacteriawere washed with PBS, fixed with an equal volume of 5% glutaraldehyde in 0.1M sodium cacodylate buffer (pH 7.4), and centrifuged at 5,000 � g for 5 min.Cells were resuspended in 1% agarose and pelleted. The bacterial pellets werestored in the fixative (2.5% glutaraldehyde) at 4°C for 2 h. The samples weredehydrated and en bloc stained with uranyl acetate in a series of ethanol con-centrations and then embedded in Epon/Araldite. Thin sections were cut usingan ultramicrotome with a diamond knife, mounted on specimen grids, andstained with uranyl acetate and lead citrate before being examined on a trans-mission electron microscope (Philips 410LS) operating at 80 kV and at magni-fications ranging from �24,000 to �55,000.

Statistical analysis. All outcome data from this study followed nonnormaldistributions. To account for this, all outcome data were ranked, and analysis ofvariance (ANOVA) or Student’s t test was then used to detect differences amongthe experimental groups. The distributions of the ranked data and the residualsfrom each ANOVA were consistent with the assumptions of procedure. If therewere more than two experimental groups in the analysis and the ANOVA was

FIG. 1. Inhibitory effect of BAL specimens from either newborn or 4- to 5-week-old piglets. Approximately 5 � 106 to 7 � 106 CFU B. pertussiswere cocultured in duplicate with SS medium (control) or BAL specimens obtained from newborn colostrum-fed, newborn colostrum-deprived,or 4- to 5-week-old piglets for 6 or 24 h. Supernatants were plated out onto BG agar plates to quantify the number of viable bacteria. Results areexpressed as the means � standard errors of the means (SEM) from two experiments with six animals per group. �, P � 0.004; ��, P � 0.002.

2340 ELAHI ET AL. INFECT. IMMUN.


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

significant, the means of the ranks were compared using Tukey’s test. Probabil-ities less than or equal to 0.05 were considered significant.

RESULTS

Inhibitory effect of BAL fluid against B. pertussis. We re-cently reported the development of a new model for pertussisin pigs (15). In this model, older piglets (4 to 5 weeks old) werefully protected against infection, whereas newborn animals (3to 5 days old) were fully susceptible to the disease. To deter-mine the mechanisms underlying protection, we investigatedthe bactericidal activity of BAL specimens obtained from non-infected piglets against B. pertussis in vitro. BAL specimensfrom either newborn (colostrum-fed/colostrum-deprived) or 4-to 5-week-old piglets were cocultured (total volume of 290 �l)with 5 � 106 to 7 � 106 CFU B. pertussis in SS medium (Fig. 1).Compared to the medium control, there was a significant re-duction in the number of viable bacteria in BAL specimensfrom 4- to 5-week-old piglets. At 6 h of incubation, bacterialnumbers were reduced by 1.5 logs (P � 0.004), and after 24 h,no viable bacteria were isolated from wells cocultured withBAL specimens obtained from 4- to 5-week-old piglets (P �0.002). In contrast, regardless of whether the BAL specimenswere obtained from colostrum-fed or colostrum-deprived new-born piglets, growth inhibition of B. pertussis was not observed(Fig. 1). Furthermore, these results confirmed that the pres-

ence of colostrum-derived secretory immunoglobulin A inBAL specimens did not have any effect on bacterial growth orinterfered with our assays. We hypothesized that BAL speci-mens obtained from older animals contained antimicrobialcomponents such as antimicrobial peptides and proteins thatcould be associated with in vivo protection against B. pertussis.

The antibacterial activity of BAL fluid against B. pertussiswas salt dependent. The antimicrobial activity of many antimi-crobial proteins and cationic peptides including beta-defensinsis greatly affected by higher salt concentrations (3). For exam-ple, it has previously been reported that an elevated concen-tration of NaCl in the airway surface fluid of cystic fibrosispatients neutralizes the antimicrobial activity of defensins andpredisposes the host to a wide range of infections (3). In orderto determine whether the observed bactericidal activity of BALspecimens in our study was due to the presence of antimicro-bial peptides, 10 to 140 mM NaCl was added to the BALspecimens. As shown in Fig. 2, 140 mM salt completely blockedand 70 mM salt partially blocked the growth-inhibitory effect ofBAL specimens against B. pertussis. However, 10 mM NaCl didnot inhibit the antibacterial activity of BAL specimens (P �

0.004) (Fig. 2). Thus, these results demonstrated that the ob-served antimicrobial effect was reversed by the addition of ahigh salt concentration to the culture and therefore furthersupported our hypothesis that the antimicrobial activity of

FIG. 2. Salt dependency of antimicrobial activity of BAL specimens. A total of 5 � 106 to 7 � 106 CFU B. pertussis were cocultured in duplicatewith SS medium (Control) or BAL specimens obtained from three 4- to 5-week-old piglets (BAL). Various concentrations of NaCl (140 mM, 70mM, and 10 mM) were added to the SS medium (Control � NaCl) or BAL specimens (BAL �NaCl) and incubated for 24 h. Supernatants wereplated onto BG agar plates to quantify the number of viable bacteria. Results are expressed as the means � SEM. �P � 0.004.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

BAL specimens obtained from older piglets could be due tothe presence of antimicrobial peptides, such as lysozyme, lac-toferrin, PR-39, PG-1, and pBD-1, in lung secretions.

Detection of pBD-1, PG-1, and PR-39 gene expression at therespiratory surfaces. Beta-defensins and cathelicidins are themost dominant antimicrobial peptides that are found in neu-trophils and epithelia as components of the early host defensesof mammals against infection. Beta-defensins are the predom-inant cationic peptides in the lung (6, 47). In fact, lower levelsof defensins may contribute to the increased susceptibility ofpremature infants to pulmonary infections (47). To determinedifferences in expression levels of pBD-1, PG-1, and PR-39between older and newborn animals, we analyzed the geneexpression of pBD-1, PG-1, and PR-39 in respiratory tissuesobtained from newborn and older piglets. Expression of pBD-1was detected in all analyzed tissues in 4- to 5-week-old piglets(Fig. 3A). In contrast, expression of pBD-1 in tissues obtainedfrom newborn piglets was detected only in the tongue epithe-lium (Fig. 3B and C). Expression of PG-1 and PR-39 wasdetected in all lung tissues obtained from either newborn orolder piglets (Fig. 4A). These results indicated that the expres-

sion of pBD-1 but not PG-1 or PR-39 developed with age.Moreover, infection with B. pertussis induced up-regulation ofpBD-1 in the lung of newborn piglets as early as 2 days post-challenge, which demonstrated that pBD-1 might be an induc-ible beta-defensin in pigs (Fig. 4B). These results were con-firmed by RP-HPLC, which revealed the presence of pBD-1 inthe BAL specimens collected from older piglets but not fromnewborn animals (Fig. 5). In addition, RP-HPLC analysis in-dicated the presence of PG-1, a highly antimicrobial peptide, inBAL specimens obtained from both newborn and older ani-mals (data not shown). Thus, these results supported our hy-pothesis that expression of pBD-1 in older piglets may contrib-ute to resistance against infection with B. pertussis.

In vitro susceptibility of B. pertussis to synthetically derivedpBD-1. We synthesized pBD-1 using Fmoc synthesis, and inorder to analyze the functional activity of this peptide, wecompared its inhibitory activity against E. coli strain JG280 tothe inhibitory activity of a recombinant baculovirus-expressedpBD-1 described previously by Shi et al. (46). Similar resultswere obtained, which demonstrated that synthetic pBD-1 wasbiologically functional. We then analyzed the inhibitory effect

FIG. 3. Tissue expression of pBD-1 mRNA in either newborn (colostrum-fed/colostrum-deprived) or 4- to 5-week-old piglets. Tissue samplescollected from (A) 4- to 5-week-old animals, (B) newborn colostrum-fed animals, or (C) newborn colostrum-deprived animals were analyzed forPCR products of 452 bp (GAPDH) and 287 bp (pBD-1) in length. Tissues included tongue (lane 1), nasal mucosa (lane 2), trachea (lane 3), lung(lane 4), intestine (lane 5), and water as a negative control (lane 6). Expression was found only in the tongue of newborn piglets but in allinvestigated tissues of 4- to 5-week-old piglets. Six animals per age group were analyzed.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

of pBD-1 against B. pertussis by inhibition assays using differentconcentrations of chemically synthesized pBD-1 in differentcultural conditions such as SS medium, PBSA (140 mM NaCl),and 20 mM PBSA (2.8 mM NaCl). First, we tested the anti-bacterial activity of pBD-1 against B. pertussis (20, 40, and 80�g/ml) in SS medium. As shown in Fig. 6A, at concentrationsof 20 �g/ml of pBD-1, the growth of B. pertussis was inhibitedby at least 2 logs after 6 h of incubation in comparison withcontrol SS medium (P 0.0001). Higher concentrations ofpBD-1 and prolonged incubation resulted in significantly in-creased inhibitory activity. Indeed, at 80 �g/ml of pBD-1 and18 h of incubation, 5 � 106 to 7 � 106 CFU of B. pertussis werecompletely neutralized (P 0.0001). However, these assayswere performed under in vitro conditions that were optimizedfor growing the bacteria. It is possible that in vivo concentra-tions of ions such as Ca2�, Mg2�, and K� as well as tissueproteins may interfere with the observed antimicrobial activity(8, 9). Although SS medium has concentrations of Ca2�,Mg2�, and K� comparable to those found in human lungsecretions, we tested the effect of the medium itself by dilutingthe SS medium in PBSA (1/10 dilution). Interestingly, pBD-1in diluted medium exhibited an even stronger bactericidal ac-

tivity against B. pertussis in a shorter period of time (P 0.0001) (Fig. 6B). Thus, pBD-1 displayed strong antimicrobialactivity against B. pertussis at concentrations of Ca2�, Mg2�,and K� that were comparable to concentrations found in hu-man lung secretions (which contain about 2 mM divalent andlesser monovalents [80 mM], mostly Na�). Moreover, the in-hibitory effect of pBD-1 against B. pertussis was also testedunder conditions (20 mM PBS) typically used for the charac-terization of AMPs. As shown in Fig. 6C, at concentrations of1 �g/ml of pBD-1, the growth of B. pertussis was inhibited byalmost 3 logs after 1 h of incubation (P 0.0001). Higherconcentrations of pBD-1 (2.5 and 5 �g/ml) completely neutral-ized 5 � 105 to 8 � 105 CFU of B. pertussis within 1 h (P 0.0001) (Fig. 6C), and prolonged incubation resulted in theabsolute neutralization of bacteria even at 1 �g/ml of pBD-1(P 0.0001) (Fig. 6C). These results demonstrated that chem-ically synthesized pBD-1 displayed strong bactericidal activityagainst B. pertussis in vitro in both a time- and dose-dependentmanner.

We also addressed the effect of various serum conditions invitro by adding 2 to 10% human, porcine, or rabbit serum tothe medium in order to simulate physiological conditions. In-

FIG. 4. Expression of pBD-1, PG-1, and PR-39 mRNA in the lungs of either newborn or 4- to 5-week-old piglets. (A) Tissue samples collectedfrom newborn or 4- to 5-week-old animals were analyzed for PCR products of 452 bp (GAPDH), 100 bp (PG-1), 262 bp (PR-39), and 287 bp(pBD-1) in length. Tissues were collected from three newborn piglets (lanes 1 to 3) and three older piglets (lanes 4 to 6), and water was used asa negative control (Neg.) (lane 7). The expression patterns of PG-1 and PR-39 were similar in both groups, but in contrast, expression of PBD-1was only found in the lungs of older piglets. (B) Expression of pBD-1 was noted in the lungs of piglets infected with B. pertussis 2 days afterchallenge but not in control animals (PBSA-challenged piglets). Results from 10 animals per age group are shown.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

FIG

.5.

RP-

HPL

Cch

rom

atog

ram

sof

BA

Lflu

idob

tain

edfr

omne

wbo

rnor

olde

rpi

glet

s.B

AL

spec

imen

sw

ere

colle

cted

in10

%ac

idic

acid

and

test

edfo

rth

epr

esen

ceof

pBD

-1.B

AL

spec

imen

sfro

mol

derp

igle

tsin

dica

ted

the

pres

ence

ofa

frac

tion

ofth

esa

me

mas

ssiz

eof

synt

hetic

pBD

-1.I

nco

ntra

st,t

hisp

eak

was

abse

ntin

BA

Lsp

ecim

enso

btai

ned

from

new

born

pigl

ets.

To

confi

rmth

epr

esen

ceof

pBD

-1,f

ract

ions

wer

eve

rifie

dby

MA

LD

Im

ass

spec

trom

etry

.mA

U,m

illi-A

bsor

banc

eU

nit.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

terestingly, after 1 h of incubation, even very high serum con-centrations (10%) did not completely inhibit the antimicrobialactivity of pBD-1 (data not shown). Thus, these results indi-cated that pBD-1 displayed biological activity in the presenceof lung secretions and serum and provided additional supportfor the hypothesis that the expression of pBD-1 at the respi-ratory surfaces was crucial for providing protection againstinfection with B. pertussis.

In vitro anti-B. pertussis activity of lysozyme, lactoferrin, andPG-1. Lysozyme and lactoferrin are the most abundant antimi-crobial factors present on mucosal surfaces of the respiratorytract (50, 52). They contribute to airway defense by exhibitingstrong antimicrobial activity against a wide range of gram-negative and gram-positive bacteria (2, 25). The antimicrobialactivity of these peptides against B. pertussis was tested by

inhibition assays using different concentrations of human ly-sozyme (100 and 1,000 mg/ml) and human lactoferrin (100 and500 mg/ml). B. pertussis was completely resistant to both pep-tides after 2 h of incubation. Higher concentrations of thesepeptides and prolonged incubation did not result in any inhib-itory effect (data not shown). In contrast, PG-1 displayedstrong anti-B. pertussis activity in vitro (complete neutralizationof 5 � 106 to 7 � 106 CFU at a concentration of 1 �g/ml for1 h). Despite strong antimicrobial activity of PG-1 against B.pertussis, it was present in the BAL specimens obtained fromboth groups of animals. These results support our hypothesisthat the observed antimicrobial activity of BAL fluid against B.pertussis was mainly due to the presence of pBD-1.

Synthetically derived pBD-1 restores in vitro activity of BALspecimens from newborn piglets against B. pertussis. To fur-

FIG. 6. Sensitivity of B. pertussis to synthetically derived pBD-1. (A) A total of 5 � 106 to 7 � 106 CFU B. pertussis were cocultured in duplicatewith different concentrations of pBD-1 (20 �g/ml, 40 �g/ml, and 80 �g/ml) in SS medium (Control) for 6, 18, and 24 h. Bacterial numbers weredetermined by plate counts. (B) A total of 5 � 106 to 7 � 106 CFU B. pertussis were cocultured in duplicate with 10 �g/ml of pBD-1 in 1/10-dilutedSS medium in PBSA for 6 or 24 h, and the number of bacteria were determined by plate counts. (C) A total of 5 � 106 to 7 � 106 CFU B. pertussiswere cocultured in duplicate with different concentrations of pBD-1 (1, 2.5, and 5 �g/ml) in 20 mM PBSA. �, all values within each time point weresignificantly different (P � 0.0001) from their respective controls. Results are expressed as the means � SEM.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

ther support our hypothesis, 2.5 �g/ml of synthesized pBD-1was added to the BAL specimens obtained from newbornpiglets (Fig. 7A). The addition of synthetic pBD-1 restored theinhibitory activity of BAL specimens from newborn animals tolevels that were comparable to those of BAL specimens fromolder animals (P 0.001). Moreover, the addition of neutral-izing antibody (rabbit anti-pBD-1 polyclonal antibody obtainedfrom rabbits immunized with synthetic pBD-1 conjugated toovalbumin) to the BAL specimens of older piglets reduced theactivity to that of BAL specimens from newborn piglets orcontrol medium (P 0.001) (Fig. 7B). In contrast, serum fromnonimmunized rabbits did not neutralize the anti-B. pertussisactivity of BAL specimens obtained from older piglets (P 0.001). Similar findings were found with synthetic peptide.While the immune serum inhibited the effects of syntheticpBD-1 against B. pertussis, the nonimmune rabbit serum hadno effect on the biological activity of this peptide (P 0.0001)(Fig. 7B). These results supported our hypothesis that theinhibitory effect of BAL fluid was specifically due to the pres-ence of pBD-1 in the BAL specimens of older piglets.

Ultrastructural analysis of B. pertussis and B. bronchisepticaafter treatment with pBD-1. To determine the potential mech-anism of action, we studied the interaction between pBD-1 andthe bacterial cell membrane using electron microscopy. Theexposure of B. pertussis in SS medium to two different concen-trations (20 and 40 �g/ml) of pBD-1 resulted in substantialmorphological damage to the cell surface of bacteria as shownby scanning electron microscopy. Untreated bacteria displayeda rough, bright surface with no apparent perforation or cellulardebris, as shown in Fig. 8. In contrast, B. pertussis exposed topBD-1 exhibited a wide range of morphological abnormalities,

as shown in Fig. 8. These included disruption of the cell sur-face, disappearance of the cell membrane, perforation andbreakage in the cell membrane, and lysis of bacteria. In con-trast, no cell structural effects were found on B. bronchisepticaeven after incubation with significantly greater concentrationsof pBD-1 (80 �g/ml).

Taken together, our in vitro studies clearly demonstrated asignificant role for pBD-1 in protection against infection withB. pertussis. To further investigate the role of pBD-1, we de-cided to test its in vivo effects against B. pertussis in newbornpiglets.

In vivo protection following treatment with pBD-1. pBD-1was delivered into the lungs of newborn piglets prior to chal-lenge with B. pertussis. We made sure that pBD-1 was com-pletely administered into the lung before the bacteria weredelivered. Indeed, treatment with only 500 �g of pBD-1 at thetime of challenge resulted in complete protection of infectedpiglets, as demonstrated by the total absence of clinical symp-toms and of pathological alterations at 2, 4, 7, and 10 dayspostinfection (Fig. 9A and B). In contrast, severe subacutehemorrhagic and necrotizing pneumonia was found in PBSA-treated piglets (Fig. 9C and D). Lesions were characterized bysevere cellular infiltrations (neutrophils and macrophages) inthe alveolar spaces, around the bronchioles, and in the walls ofblood vessels with severe alveolar hemorrhagic, congestion,edema, and focal bronchiolar necrosis causing suppurative andhistiocytic pleuropneumonia and bronchointerstitial pneumo-nia (data not shown). In addition, numbers of isolated bacteriain BAL specimens collected from pBD-1-treated animals weresignificantly reduced at 2, 4, and 7 days postinfection (P 0.0001) (Fig. 10A). In addition, significantly lower numbers of

FIG. 7. Restoration of the antimicrobial activity of BAL specimens obtained from newborn piglets by synthetic pBD-1 and neutralizing effectsof anti-pBD-1 antibody. (A) A total of 1 � 106 CFU B. pertussis were cocultured with BAL specimens obtained from newborn piglets with (BAL �pBD-1) (2.5 �g/ml) or without (BAL) synthetic pBD-1 for 2 h. Controls include SS medium (Control) and synthetic pBD-1 (2.5 �g/ml). (B) A totalof 1 � 106 CFU B. pertussis were cocultured with BAL specimens or synthetic pBD-1 plus rabbit immune (1%) or nonimmune (1%) serumfor 3 h. Columns include SS medium (Control), BAL specimens obtained from older animals (BAL), BAL specimens obtained from older animalsplus immune serum (BAL � immune serum) or nonimmune serum (BAL � non-immune serum), synthetic pBD-1 plus immune serum (pBD-1 �immune serum), and synthetic pBD-1 plus nonimmune serum (pBD-1 � non-immune serum). Bacterial numbers were determined by plate counts. Dataare presented as the means � SEM. �P 0.001.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

bacteria were found in homogenized lung tissues of treatedanimals at 2, 4, 7, and 10 days postinfection (P 0.005) (Fig.10B) compared to BAL specimens and tissues from controlanimals, which contained between 105 and 107 CFU/ml and 105

and 108 CFU/g, respectively, at these days. Thus, our resultsdemonstrated that defensins played an essential role in medi-ating protection against pertussis. Furthermore, our resultsclearly demonstrate that treatment with chemically synthesizedHDPs may represent a possible means of treating this impor-tant disease in young children and infants.

pBD-1 enhances innate immunity in vivo. To further inves-tigate whether the observed protection was mediated primarilyby direct antimicrobial activity or stimulation of innate immu-nity, we deposited 500 �g of pBD-1 in 1.5 ml PBSA (20 mM)into lungs of newborn piglets 4 h prior to challenge. Controlanimals received the same volume of PBSA instead. Both

groups of piglets were challenged 4 h later with 5 � 109 CFUbacteria. Treatment with pBD-1 4 h prior to challenge resultedin a reduction in lesion size and bacterial load compared tocontrol animals (data not shown). Furthermore, between 1.2 �102 and 9.5 � 102 CFU of B. pertussis/ml BAL fluid and 1.6 �102 to 8.8 � 102 CFU of B. pertussis /g tissue were isolated fromtreated piglets at days 2 and 4 postchallenge, whereas between2.5 � 102 and 3.5 � 104 CFU/ml BAL fluid and 1.6 � 103 and4.4 � 103 CFU/g tissue were isolated from control animals onthese days, respectively (data not shown). These differenceswere highly significant for the lesions (P 0.0001). Thus, theseresults indicate that protection was mediated, at least in part,through stimulation of the innate immune response.

Specificity of antimicrobial peptides. HDPs represent po-tential new drug candidates for innate immune treatments inhumans and animals. However, resistance against HDPs has

FIG. 8. Microscopical alterations after exposure to pBD-1. A total of 5 � 109 to 7 � 109 CFU of B. pertussis and B. bronchiseptica were exposedto pBD-1. (A) B. pertussis cells left untreated; (B) B. pertussis cells treated with 20 �g/ml of pBD-1 (B1, B2, and B3 represent different time pointsincluding 20, 40, and 60 min, respectively); (C) B. bronchiseptica left untreated; (D) B. bronchiseptica treated with 80 �g/ml of pBD-1. Micrographsare representative of at least 10 examined microscopic fields per sample. Magnification, �40,500.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

already been described for a number of bacteria (11, 18, 22, 33,37, 39, 49). The present disease model allows us to compare anatural pathogen of pigs (B. bronchiseptica) to an experimen-tally introduced human pathogen (B. pertussis). We thereforecompared the bactericidal activity of pBD-1 and PG-1 againstboth B. pertussis and B. bronchiseptica in vitro using 5 and 10�g/ml of pBD-1 and 1, 5, and 10 �g/ml of PG-1. Interestingly,while �5 �g/ml pBD-1 inhibited the growth of B. pertussis by100% after 1 h, the growth of B. bronchiseptica was unaffectedat this time point (P 0.001) (Fig. 11). In addition, B. bron-chiseptica was resistant to pBD-1 even at very high concentra-

tions (50 �g/ml) and for up to 6 h of incubation (data notshown). PG-1 exhibited very strong antimicrobial activityagainst B. pertussis, with complete elimination of 5 � 106 to7 � 106 CFU in 1 h. In contrast, B. bronchiseptica was resistantto this peptide (data not shown). Thus, these results indicatedthat B. bronchiseptica, a natural pathogen of pigs, has evolvedmeans of evading the porcine innate immune system. Wetherefore measured the bactericidal activity of hBD-2, the hu-man defensin that is most homologous to pBD-1, against bothB. pertussis and B. bronchiseptica. While hBD-2 had no bacte-ricidal activity against B. bronchiseptica in vitro at concentra-

FIG. 9. Gross pathology of infected and treated lungs. Shown are lungs from piglets infected with 5 � 109 CFU of B. pertussis and treated with500 �g pBD-1 at (A) 2 days and (B) 4 days postchallenge or untreated infected control (saline-treated) piglets at (C) 2 and (D) 4 days afterchallenge. Results shown are representative of three experiments, with at least three piglets per time point.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

tions of 5 and 10 �g/ml, it exhibited some inhibitory effectagainst B. pertussis in the same period of time (P 0.001) (Fig.11). However, hBD-2 (20 �g/ml) was not able to eliminate100% of the bacteria even after 4 h of incubation (data notshown). These results suggest that both B. pertussis and B. bron-chiseptica have evolved in their respective host environments andthat both have developed strategies to evade the innate immunedefense. Thus, the development of potential resistance againstHDPs represents a challenge for the development of future in-nate immune treatments. Clearly, more studies are required todetermine the mechanisms of resistance.

In summary, our results demonstrate that the presence ofpBD-1 was associated with complete protection against respi-ratory infection with B. pertussis. Furthermore, these resultsdemonstrate that pathogens may have developed novel strat-egies for overcoming recognition by the innate immune systemin their natural host environments.

DISCUSSION

Pertussis represents an important respiratory disease of in-fants and young children and still accounts for significant num-bers of morbidity and mortality worldwide. Here, we show thatthe presence of pBD-1, a porcine defensin homologue ofhBD-2, was associated with protection against infection with B.pertussis. Our results demonstrate the great potential of HDPsas an alternative to antibiotics for the treatment of life-threat-ening infections in young children and infants. Moreover, ourresults demonstrate the importance of innate immunity in thepathogenesis of pertussis.

In contrast to newborn piglets, older piglets were fully pro-tected against infection with B. pertussis and, in fact, very rap-idly cleared infection with 5 � 109 CFU of B. pertussis within24 h. These results clearly demonstrated the importance ofinnate immunity for protecting the pig against infection with B.

pertussis. Similarly, adult humans or mice typically displaymuch milder clinical symptoms of pertussis than young infantsor neonatal mice, which often succumb to an infection. Thus,a mature and fully developed innate immune system seems tobe essential for surviving infection with B. pertussis. Indeed,additional evidence to support the importance of innate im-munity came from recent studies using mice, which demon-strated that the secretion of gamma interferon at the upperrespiratory surfaces following infection with B. pertussis wasresponsible for confining the bacteria to the lungs and thesubsequent activation of infected macrophages (51). Further-more, complement-mediated killing and neutralization by an-timicrobial peptides and surfactant proteins A and D (pulmo-nary collectins) are thought to represent important mechanismsfor preventing infections with bordetellae (17, 38, 43). Al-though most of these studies were performed in vitro, together,they indicate the importance of innate immunity in confiningand preventing infections with B. pertussis. However, the ef-fectiveness of innate immunity is often restricted by the devel-opment of bacterial resistance. For example, the B. pertussislipopolysaccharide molecule is thought to shield bacteria fromneutralization by the pulmonary collectins Sp-A and Sp-D orcomplement-mediated killing (38, 43). Thus, more studies arerequired to determine the relative importance of innate immu-nity against pertussis.

In the present study, we demonstrated that protection wasassociated with the presence of pBD-1 at the upper respiratorysurfaces. pBD-1 was previously described to be constitutivelyexpressed throughout various porcine tissues (61). However,infection with B. pertussis induced a significant up-regulation ofpBD-1 in the lungs of newborn piglets, which is consistent withobservations made previously by Sang et al. (41) that indicatedthat pBD-1 may in fact represent an inducible defensin. In thepresent study, pBD-1 transcripts were detected in tongue, in-

FIG. 10. Effect of in vivo treatment with pBD-1 on the bacterial load in the lung. (A) At each time point, BAL specimens from two to threepiglets per saline-treated group (Control) or pBD-1-treated (Treated) group were plated onto BG agar plates to determine the number of viablebacteria within the BAL fluid. Data are expressed as the means � SEM. �P 0.001. (B) Macroscopically altered tissues were collected, weighed,homogenized, and plated onto BG plates to determine bacterial counts. Data are expressed as the means � SEM. �P 0.005.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

testine, nasal mucosa, trachea, and lung tissues of older piglets.In contrast, in newborn animals, pBD-1 gene expression wasrestricted to the tongue epithelium, with no transcripts de-tected in the upper respiratory tract. This observation wasconfirmed by RP-HPLC analysis, which revealed the presenceof pBD-1 only in BAL specimens collected from older animals.This suggested that lower levels of pBD-1 contributed to theincreased disease susceptibility of newborn piglets. Indeed,differential regulation of HDPs has been correlated with var-ious disease outcomes in several disease models. Whereas up-regulation of HDPs was thought to represent a marker forprotection, down-regulation or even the absence of HDPs of-ten leads to full development of disease (23, 34). For example,murine beta-defensin 1-deficient mice demonstrate a delayedclearance of Haemophilus influenzae from the lungs followingrespiratory infection (23, 32, 34). Moreover, human patientswith alpha-defensin deficiency or lower levels of histatin pep-tides suffer from recurrent and severe bacterial and fungalinfections (1, 23). Thus, our results are consistent with theconcept that secreted HDPs constitute the first line of defenseagainst respiratory pathogens (45).

HDPs have a variety of immunological functions, with direct

antimicrobial activity being only one of them. For example,HDPs act as signaling molecules that can activate host cellularprocesses that are involved in immune defense as well aswound healing (9, 20, 24). It is becoming more evident thatmammalian HDPs have strong immunostimulatory and immu-nomodulatory activities, including the ability to stimulate che-motaxis of immature DCs and T cells, activation of antigen-presenting cells, and induction of glucocorticoid production,macrophage phagocytosis, mast cell degranulation, comple-ment activation, and IL-8 production by epithelial cells (7, 17,36, 56, 57). Thus, HDPs appear to represent an important linkbetween innate and acquired immunity and are potent immunemodulators and adjuvants for vaccines. In addition to the largenumber of in vitro studies, many studies have demonstratedthe therapeutic potential of HDPs for treating infectious dis-eases in vivo. Typically, in these studies, HDP expression levelswere increased either by injection of recombinant or synthe-sized HDPs or by transgenic overexpression of the HDP. Forexample, intratracheal instillation or systemic injection of LL-37, a peptide derived from the human cathelicidin hCAP-18,resulted in a significant reduction of the bacterial load andinflammatory response and improved survival rates following

FIG. 11. Susceptibility of B. pertussis and B. bronchiseptica to pBD-1 and hBD-2. A total of 5 � 106 to 7 � 106 CFU/ml of B. pertussis and B.bronchiseptica were exposed to 5 and 10 �g/ml of pBD-1 (gray columns) and 5 or 10 �g/ml of hBD-2 (hatched column) or grown in SS medium(black column) for 1 h. The number of bacteria were determined by plate counts. �, P � 0.0001. Data are presented as the means � SEM.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

pulmonary challenge with Pseudomonas aeruginosa (5). Similarresults were obtained following intravenous injection withEscherichia coli (5). In sheep, pulmonary administration of thesheep macrophage antimicrobial peptide 29 (SMAP-29) re-sulted in protection against acute pneumonia caused byMannheimia hemolytica (10). Furthermore, transgenic expres-sion of hBD-5 in mouse intestinal Paneth cells led to protec-tion against a lethal challenge with Salmonella enterica serovarTyphimurium (40). The possible mechanism by which pBD-1confers protection against B. pertussis in the present study isnot yet clear. First, in view of the very small amounts of pBD-1(only 500 �g), it is extremely unlikely that the observed effectscan be explained solely by a direct antibacterial effect of de-posited pBD-1. Second, by changing the pulmonary tube be-tween injections, we ensured that pBD-1 and bacteria did notcome into direct contact before delivery into the lung. Third,preliminary studies of mice following intranasal deposition ofpBD-1 indicated the recruitment of neutrophils and mononu-clear cells around bronchioles and into the airways as early as6 h after deposition (data not shown). These observationssuggest that leukocytes that have been recruited as a result ofpBD-1 treatment may clear the local infection. In agreementwith this view, Welling et al. previously demonstrated in anexperimental peritoneal and thigh muscle bacterial infectionmodel that human neutrophil peptide 1 displayed marked invivo antibacterial activity in mice and that this activity ap-peared to be mediated by local leukocyte accumulation (55).Also, preliminary studies in our laboratory using porcinePBMCs and lymph node cells demonstrated up-regulation oftypical Th1-type cytokines such as IL-12 and gamma interferonafter in vitro stimulation with pBD-1 (data not shown). More-over, pretreatment with pBD-1 4 h prior to challenge infectionresulted in smaller lesions and significantly lower bacterialnumbers. pBD-1 significantly diminished the number of B.pertussis CFU recovered from treated animals compared withthose recovered from saline-treated piglets. These findings in-dicate that protection was mediated, at least in part, by stim-ulation of the innate immune response. In this particular study,however, lower bacterial numbers were recovered from saline-treated animals than in previous studies, which might be due tothe repeat anesthesia and intubation procedures. Further stud-ies are required to explain whether stress related to thesecomplex procedures can reduce the number of bacteria in thelung. Our findings suggest that protection of pigs against B.pertussis may be mediated by a combination of mechanismsincluding direct antimicrobial activity as well as immunostimu-latory activity. Further studies are required to determine theoptimal peptide concentration and time point to enhance theinnate immune response against infection.

B. pertussis and B. bronchiseptica are respiratory tract patho-gens that are derived from the same ancestor. However, whileB. pertussis is an obligate human pathogen that also infectsmice and pigs under experimental conditions, B. bronchisepticahas a broader host spectrum that includes birds, mammals, andimmunocompromised people (21). Interestingly, althoughclosely related, both pathogens exhibited a very different pat-tern of resistance against pBD-1 and hBD-2. While B. bron-chiseptica, a pathogen of pigs and humans, was fully resistant toboth pBD-1 and hBD-2, B. pertussis was highly susceptible toneutralization by pBD-1 but not fully resistant to hBD-2. Thus,

these results indicate that these two bacteria have evolved intheir respective microenvironments and have developed mech-anisms to overcome recognition by the respective innate im-mune systems. This observation clearly challenges the under-standing of HDPs, which suggests that neutralization ofbacteria occurs in an unspecific fashion. Furthermore, thisobservation is of great importance for future development ofHDP-based immune therapeutics, since it demonstrates thatAMP activity may be relatively restricted to individual bacteria.To address this hypothesis, we are currently testing a numberof bacterial mutants that display altered lipopolysaccharidemolecules in order to identify the interactions between pBD-1and the bacterial membrane.

In summary, we demonstrated that in vivo treatment withpBD-1 conferred protection against respiratory infection withB. pertussis in newborn piglets. Our results therefore demon-strate the importance of innate immunity against pertussis andindicate that HDPs represent an alternative to antibiotics.Moreover, our results also indicate that the mechanisms ofprotection may be a combination of both direct antimicrobialactivity and immunostimulatory activity and that bacteriaevolve to the specific microenvironments in their respectivehosts.

ACKNOWLEDGMENTS

We are thankful to the Vaccine and Infectious Disease OrganizationAnimal Care Staff for their invaluable assistance with housing, challeng-ing, and monitoring of animals, especially Kuldip Mirakhur, AmandaGiesbrecht, Sherry Tetland, Lucas Wirth, and Stacy Miskolcszi.

This study was funded by a grant from the Bill & Melinda GatesFoundation and the Canadian Institutes of Health Research throughthe Grand Challenges in Global Health Initiative (http://www.grand-challengesgh.org/), the Canadian Institute of Health Research(CIHR), the Krembil Foundation, and the Saskatchewan Health Re-search Foundation (SHRF; postdoctoral fellowship to S. Elahi).

This article was published with permission of the Director of theVaccine and Infectious Disease Organization as article number 399.

REFERENCES

1. Andreu, D., and L. Rivas. 1998. Animal antimicrobial peptides: an overview.Biopolymers 47:415–433.

2. Arnold, R. R., M. Brewer, and J. J. Gauthier. 1980. Bactericidal activity ofhuman lactoferrin: sensitivity of a variety of microorganisms. Infect. Immun.28:893–898.

3. Bals, R., M. J. Goldman, and J. M. Wilson. 1998. Mouse beta-defensin 1 isa salt-sensitive antimicrobial peptide present in epithelia of the lung andurogenital tract. Infect. Immun. 66:1225–1232.

4. Bals, R., and P. S. Hiemstra. 2004. Innate immunity in the lung: howepithelial cells fight against respiratory pathogens. Eur. Respir. J. 23:327–333.

5. Bals, R., D. J. Weiner, A. D. Moscioni, R. L. Meegalla, and J. M. Wilson.1999. Augmentation of innate host defense by expression of a cathelicidinantimicrobial peptide. Infect. Immun. 67:6084–6089.

6. Beisswenger, C., and R. Bals. 2005. Antimicrobial peptides in lung inflam-mation. Chem. Immunol. Allergy 86:55–71.

7. Biragyn, A., I. M. Belyakov, Y. H. Chow, D. S. Dimitrov, J. A. Berzofsky, andL. W. Kwak. 2002. DNA vaccines encoding human immunodeficiency virus-1glycoprotein 120 fusions with proinflammatory chemoattractants induce sys-temic and mucosal immune responses. Blood 100:1153–1159.

8. Boman, H. G. 2003. Antimicrobial peptides and their role in innate immu-nity. J. Intern. Med. 254:197–215.

9. Bowdish, D. M., D. J. Davidson, and R. E. Hancock. 2005. A re-evaluation ofthe role of host defence peptides in mammalian immunity. Curr. ProteinPept. Sci. 6:35–51.

10. Brogden, K. A., V. C. Kalfa, M. R. Ackermann, D. E. Palmquist, P. B.McCray, Jr., and B. F. Tack. 2001. The ovine cathelicidin SMAP29 killsovine respiratory pathogens in vitro and in an ovine model of pulmonaryinfection. Antimicrob. Agents Chemother. 45:331–334.

11. Campos, M. A., M. A. Vargas, V. Regueiro, C. M. Llompart, S. Albertı́, andJ. A. Bengoechea. 2004. Capsule polysaccharide mediates bacterial resistanceto antimicrobial peptides. Infect. Immun. 72:7107–7114.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

12. Cotter, P. A., and J. F. Miller. 1994. BvgAS-mediated signal transduction:analysis of phase-locked regulatory mutants of Bordetella bronchiseptica in arabbit model. Infect. Immun. 62:3381–3390.

13. Crowcroft, N. S., C. Stein, P. Duclos, and M. Birmingham. 2003. How bestto estimate the global burden of pertussis? Lancet Infect. Dis. 3:413–418.

14. Diamond, G., J. P. Russell, and C. L. Bevins. 1996. Inducible expression ofan antibiotic peptide gene in lipopolysaccharide-challenged tracheal epithe-lial cells. Proc. Natl. Acad. Sci. USA 93:5156–5160.

15. Elahi, S., R. Brownlie, J. Korzeniowski, R. Buchanan, B. O’Connor, M. S.Peppler, L. A. Babiuk, S. A. Halperin, S. F. Lee, and V. Gerdts. 2005.Infection of newborn piglets with Bordetella pertussis: a new model for per-tussis. Infect. Immun. 73:3636–3645.

16. Fahlgren, A., S. Hammarstrom, A. Danielsson, and M. L. Hammarstrom.2004. Beta-defensin-3 and -4 in intestinal epithelial cells display increasedmRNA expression in ulcerative colitis. Clin. Exp. Immunol. 137:379–385.

17. Fernandez, R. C., and A. A. Weiss. 1996. Susceptibilities of Bordetella per-tussis strains to antimicrobial peptides. Antimicrob. Agents Chemother. 40:1041–1043.

18. Foxman, B., L. Zhang, K. Palin, P. Tallman, and C. F. Marrs. 1995. Bacterialvirulence characteristics of Escherichia coli isolates from first-time urinarytract infection. J. Infect. Dis. 171:1514–1521.

19. Frohm, M., B. Agerberth, G. Ahangari, M. Stahle-Backdahl, S. Liden, H.Wigzell, and G. H. Gudmundsson. 1997. The expression of the gene codingfor the antibacterial peptide LL-37 is induced in human keratinocytes duringinflammatory disorders. J. Biol. Chem. 272:15258–15263.

20. Gallo, R. L., M. Murakami, T. Ohtake, and M. Zaiou. 2002. Biology andclinical relevance of naturally occurring antimicrobial peptides. J. AllergyClin. Immunol. 110:823–831.

21. Giles, C. J. 1992. Bordetellosis. Iowa State University Press, Ames, Iowa.22. Guina, T., E. C. Yi, H. Wang, M. Hackett, and S. I. Miller. 2000. A PhoP-

regulated outer membrane protease of Salmonella enterica serovar Typhi-murium promotes resistance to alpha-helical antimicrobial peptides. J. Bac-teriol. 182:4077–4086.

23. Hancock, R. E., and D. S. Chapple. 1999. Peptide antibiotics. Antimicrob.Agents Chemother. 43:1317–1323.

24. Hancock, R. E. W., and G. Diamond. 2000. The role of cationic antimicrobialpeptides in innate host defences. Trends Microbiol. 8:402–410.

25. Harbitz, O., A. O. Jenssen, and O. Smidsrod. 1984. Lysozyme and lactoferrinin sputum from patients with chronic obstructive lung disease. Eur. J. Respir.Dis. 65:512–520.

26. Harder, J., J. Bartels, E. Christophers, and J. M. Schroder. 2001. Isolationand characterization of human beta-defensin-3, a novel human induciblepeptide antibiotic. J. Biol. Chem. 276:5707–5713.

27. Harder, J., R. Siebert, Y. Zhang, P. Matthiesen, E. Christophers, B. Schlegel-berger, and J. M. Schroder. 1997. Mapping of the gene encoding human beta-defensin-2 (DEFB2) to chromosome region 8p22-p23.1. Genomics 46:472–475.

28. Harlow, E., and D. Lane. 1988. Antibodies: a laboratory manual. Cold SpringHarbor Laboratory, Cold Spring Harbor, N.Y.

29. Heimstra, P. S., B. A. Fernie-King, J. McMichael, P. J. Lachmann, and J.Sallenave. 2004. Antimicrobial peptides: mediators of innate immunity astemplate for the development of novel anti-infective and immune therapy.Curr. Pharm. Des. 10:2891–2905.

30. Hiratsuka, T., M. Nakazato, Y. Date, J. Ashitani, T. Minematsu, N. Chino,and S. Matsukura. 1998. Identification of human beta-defensin-2 in respi-ratory tract and plasma and its increase in bacterial pneumonia. Biochem.Biophys. Res. Commun. 249:943–947.

31. Izadpanah, A., and R. L. Gallo. 2005. Antimicrobial peptides. J. Am. Acad.Dermatol. 52:381–392.

32. Lee, H., A. Andalibi, P. Webster, S. K. Moon, K. Teufert, S. H. Kang, J. D.Li, M. Nagura, T. Ganz, and D. J. Lim. 2004. Antimicrobial activity of innateimmune molecules against Streptococcus pneumoniae, Moraxella catarrhalisand nontypeable Haemophilus influenzae. BMC Infect. Dis. 4:4–12.

33. Legarda, D., M. E. Klein-Patel, S. Yim, M. H. Yuk, and G. Diamond. 2005.Suppression of NF-kappaB-mediated beta-defensin gene expression in themammalian airway by the Bordetella type III secretion system. Cell. Micro-biol. 7:489–497.

34. Moser, C., D. J. Weiner, E. Lysenko, R. Bals, J. N. Weiser, and J. M. Wilson.2002. -Defensin 1 contributes to pulmonary innate immunity in mice. In-fect. Immun. 70:3068–3072.

35. Niyonsaba, F., H. Ogawa, and I. Nagaoka. 2004. Human -defensin-2 func-tions as a chemotactic agent for tumor necrosis factor-�-treated humanneutrophils. Immunology 111:273–281.

36. Oppenheim, J. J., A. Biragyn, L. W. Kwak, and D. Yang. 2003. Roles ofantimicrobial peptides such as defensins in innate and adaptive immunity.Ann. Rheum. Dis. 62(Suppl. 2):ii17–ii21.

37. Peschel, A., and L. V. Collins. 2001. Staphylococcal resistance to antimicro-bial peptides of mammalian and bacterial origin. Peptides 22:1651–1659.

38. Pishko, E. J., D. J. Betting, C. S. Hutter, and E. T. Harvill. 2003. Bordetellapertussis acquires resistance to complement-mediated killing in vivo. Infect.Immun. 71:4936–4942.

39. Rouquette-Loughlin, C., S. A. Dunham, M. Kuhn, J. T. Balthazar, and W. M.Shafer. 2003. The NorM efflux pump of Neisseria gonorrhoeae and Neisseriameningitidis recognizes antimicrobial cationic compounds. J. Bacteriol. 185:1101–1106.

40. Salzman, N. H., D. Ghosh, K. M. Huttner, Y. Paterson, and C. L. Bevins.2003. Protection against enteric salmonellosis in transgenic mice expressinga human intestinal defensin. Nature 422:522–526.

41. Sang, Y., B. Ramanathan, C. R. Ross, and F. Blecha. 2005. Gene silencingand overexpression of porcine peptidoglycan recognition protein long iso-forms: involvement in -defensin-1 expression. Infect. Immun. 73:7133–7141.

42. Reference deleted.43. Schaeffer, L. M., F. X. McCormack, H. Wu, and A. A. Weiss. 2004. Interac-

tions of pulmonary collectins with Bordetella bronchiseptica and Bordetellapertussis lipopolysaccharide elucidate the structural basis of their antimicro-bial activities. Infect. Immun. 72:7124–7130.

44. Schonwetter, B. S., E. D. Stolzenberg, and M. A. Zasloff. 1995. Epithelialantibiotics induced at sites of inflammation. Science 267:1645–1648.

45. Schutte, B. C., and P. B. J. McCray. 2002. Beta-defensins in lung hostdefense. Annu. Rev. Physiol. 64:709–748.

46. Shi, J., G. Zhang, H. Wu, C. Ross, F. Blecha, and T. Ganz. 1999. Porcineepithelial -defensin 1 is expressed in the dorsal tongue at antimicrobialconcentrations. Infect. Immun. 67:3121–3127.

47. Starner, T. D., B. Agerberth, G. H. Gudmundsson, and P. B. McCray, Jr.2005. Expression and activity of beta-defensins and LL-37 in the developinghuman lung. J. Immunol. 174:1608–1615.

48. Steinberg, D. A., M. A. Hurst, C. A. Fujii, A. H. Kung, J. F. Ho, F. C. Cheng,D. J. Loury, and J. C. Fiddes. 1997. Protegrin-1: a broad-spectrum, rapidlymicrobicidal peptide with in vivo activity. Antimicrob. Agents Chemother.41:1738–1742.

49. Stumpe, S., R. Schmid, D. L. Stephens, G. Georgiou, and E. P. Bakker. 1998.Identification of OmpT as the protease that hydrolyzes the antimicrobialpeptide protamine before it enters growing cells of Escherichia coli. J. Bac-teriol. 180:4002–4006.

50. Thompson, A. B., T. Bohling, F. Payvandi, and S. I. Rennard. 1990. Lowerrespiratory tract lactoferrin and lysozyme arise primarily in the airways andare elevated in association with chronic bronchitis. J. Lab. Clin. Med. 115:148–158.

51. Torre, D., A. Pugliese, F. Speranza, G. P. Fiori, L. Perversi, P. Marone, andR. Tambini. 1993. Interferon-gamma levels in serum and bronchoalveolarlavage fluid of mice infected with Bordetella pertussis. J. Infect. Dis. 167:762–765.

52. Travis, S. M., B. A. Conway, J. Zabner, J. J. Smith, N. N. Anderson, P. K.Singh, E. P. Greenberg, and M. J. Welsh. 1999. Activity of abundant anti-microbials of the human airway. Am. J. Respir. Cell Mol. Biol. 20:872–879.

53. Valore, E. V., C. H. Park, A. J. Quayle, K. R. Wiles, P. B. J. McCray, and T.Ganz. 1998. Human beta-defensin-1: an antimicrobial peptide of urogenitaltissues. J. Clin. Investig. 101:1633–1642.

54. Vora, P., A. Youdim, L. S. Thomas, M. Fukata, S. Y. Tesfay, K. Lukasek,K. S. Michelsen, A. Wada, T. Hirayama, M. Arditi, and M. T. Abreu. 2004.Beta-defensin-2 expression is regulated by TLR signaling in intestinal epi-thelial cells. J. Immunol. 173:5398–5405.

55. Welling, M. M., P. S. Hiemstra, M. T. van den Barselaar, A. Paulusma-Annema, P. H. Nibbering, E. K. Pauwels, and W. Calame. 1998. Antibacte-rial activity of human neutrophil defensins in experimental infections in miceis accompanied by increased leukocyte accumulation. J. Clin. Investig. 102:1583–1590.

56. Yang, D., A. Biragyn, M. D. Hoover, J. Lubkowski, and J. J. Oppenheim.2004. Multiple roles of antimicrobial defensins, cathelicidins, and eosinophil-derived neurotoxin in host defense. Annu. Res. Immunol. 22:181–215.

57. Yang, D., O. Chertov, S. N. Bykovskaia, Q. Chen, M. J. Buffo, J. Shogan, M.Anderson, J. M. Schroder, J. M. Wang, O. M. Howard, and J. J. Oppenheim.1999. Beta-defensins: linking innate and adaptive immunity through den-dritic and T cell CCR6. Science 286:525–528.

58. Yeaman, M. R., and Y. Y. Nannette. 2003. Mechanisms of antimicrobialpeptide action and resistance. Pharmacol. Rev. 55:27–55.

59. Zanetti, M. 2004. Cathelicidins, multifunctional peptides of the innate im-munity. J. Leukoc. Biol. 75:39–48.

60. Zasloff, M. 2002. Antimicrobial peptides of multicellular organisms. Nature415:389–395.

61. Zhang, G., H. Wu, J. Shi, T. Ganz, C. R. Ross, and F. Blecha. 1998. Molec-ular cloning and tissue expression of porcine beta-defensin-1. FEBS Lett.424:37–40.

Editor: A. D. O’Brien



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

The Host Defense Peptide Beta-Defensin 1 Confers Protection against Bordetella pertussis in Newborn...

Documents

Transcript of The Host Defense Peptide Beta-Defensin 1 Confers Protection against Bordetella pertussis in Newborn...